[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [wsrp-primer] Re: [wsrp-markup] Re: [wsrp-primer] A consistent use-case for primer
Rex, I respectfully disagree as I've not had a chance to follow the debate on this. I'm not saying that we should or should not change anything, just asking for time on this, as it makes it really hard for me to to change right now on the working draft in track mode. It is easier for me to change latter. Regards, Subbu Rex Brooks said the following on 10/27/2003 09:55 AM: > Two points, Subbu, > > 1. If you're using a scenario that we decide against in the end, you > may end up having to rework what you are writing now. > > 2. If we decide on the White Paper Scenario, we can put that issue to > bed, alleviating 1. Then we don't face a situation where we have hours > of work at risk. Having just done about 14 hours, I really do > understand that aspect, and that is why I did not make the work I did > over the weekend dependent on a scenario per se. > > There's only four of us, and we already have 2 in favor of White Paper. > > Alan, do you have some thoughts on this? > > Ciao, > Rex > > At 7:33 AM -0700 10/27/03, Subbu Allamaraju wrote: > >> Rich/Rex, >> >> Could we hold on a bit so that the drafts can continue? For the parts >> I'm writing, I'm constantly referring to the scenario (here is what >> the consumer/producer is trying to do, and here is the message and so >> on), and it would be convenient to wait till we've a better picture >> for the flow and not change anything for now. >> >> Regards, >> >> Subbu >> >> Rich Thompson said the following on 10/27/2003 07:24 AM: >> >>> >>> I am sensitive to the issue you raise. I find that the criteria for >>> a good example scenario are: >>> >>> 1. Involved enough to demonstrate the desired functionality >>> 2. Simple enough that understanding the example does not detract >>> from understanding the functionality demonstrated. >>> 3. Showy enough that it holds people's attention while they hear >>> about the functionality. >>> >>> Finding a good tradeoff between these three is never trivial. Easy >>> traps to fall into include: >>> 1. A complex example that people outside of its domain never get >>> their head around (and therefore miss what is being demonstrated). >>> 2. A "nothing new" example that causes people to ignore what is >>> being said. >>> >>> I think you are concerned that the current scenario falls into trap >>> #2 while I am encouraging us to be careful for trap #1 as well. >>> >>> Thinking briefly about this, why not use the scenario from the >>> Whitepaper? It is both simple enough for people to rapidly grasp and >>> everyday enough for people to understand the value of having some of >>> the scenario being local (i.e. intra-company if not on the portal >>> server) and some remote. It demonstrates the value of the >>> functionality without getting into the issues of "aren't there >>> better ways than what you demoed in your portlets?" that Rex is >>> raising for the financial scenario. It also would tie the Primer to >>> what we would consider the more introductory material on WSRP, and >>> in fact the Primer can refer back to the White paper as providing >>> such material. >>> >>> Rich >>> >>> >>> *Rex Brooks <rexb@starbourne.com>* >>> >>> 10/25/2003 08:55 AM >>> >>> >>> To >>> Rich Thompson/Watson/IBM@IBMUS, >>> wsrp-markup@lists.oasis-open.org, wsrp-primer@lists.oasis-open.org >>> cc >>> >>> Subject >>> [wsrp-markup] Re: [wsrp-primer] A consistent use-case for primer >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> Thanks, Rich, >>> >>> This is okay, but I really don't have the time to develop a second >>> scenario simply because it is more advanced. Nor do I have the time >>> and energy to put into attempting to persuade the group that this is >>> a significant enough issue to justify the extra work, but I will >>> take a last stab at it just in case I did not make a clear enough >>> case. Further, the reason why I think I was not clear enough is that >>> I had only seen the TV spots the previous evening. That is why I did >>> not have a clear alternative in mind--because I had not had enough >>> time to absorb the implications the TV ad presented in order to >>> express my concerns more cogently. The article Russell and I wrote >>> (about which I informed you 10/17) stimulated the presentaton >>> opportunity which has occupied me much of this week, and my >>> preoccupation has delayed this absorption. >>> >>> The point is not so much that the stock brokerage example is old and >>> familiar so much as the fact that some aspects of the use case are >>> being superceded in the marketplace by products on which advertising >>> dollars are being spent on media buys in the national broadcast and >>> cable television markets.This investment is being spent specifically >>> on "streaming" online stock market information. >>> >>> My point is also that this kind of media spending in this economy >>> means that people working in those marketplaces (media and stock >>> markets) will be aware of it, as will IT firms which sell into those >>> markets. >>> >>> I am not saying that the developers/implementers will be aware of >>> it, nor that it will enter into their concerns, but that it enter >>> into the concerns of the IT decision makers on whether to set the >>> implementers to work using WSRP/JSR168. >>> >>> Whether the IT decision-makers and the IT implementers discuss their >>> respective viewpoints is difficult to establish, and not worth >>> basing our decision solely on such an unfounded assumption. It is >>> even more difficult to guess if they include these kinds of >>> discussions. However, we need to understand the ramifications if >>> they do. Would we consider implementing a standard that uses an >>> primer example that is apparently being superceded in the media >>> marketplace by spending on media buys for apparently newer and >>> faster information delivery systems. >>> >>> I know that is a lot of imponderable iffiness, in this, and that is >>> why I am not going to start any work on the basis of guessing if my >>> points persuade anyone, but continue working on what we have, not on >>> my suggestions. >>> >>> However, as an advertising person, I would be making the case to us, >>> if we were my client, to use an example that does not run the risk >>> of being considered passé in the perceptions of_ IT decision-makers, >>> not implementers_, since, except in software engineering firms, it >>> is not usually the implementers who make these decisions. >>> >>> Regardless, as convoluted as this is, I did want to attempt to make >>> it somewhat more clear. It boils down to making our standard clearly >>> cutting edge in the perceptions of our marketplace--the IT >>> decision-makers, be they software engineers or software engineering >>> company executives. >>> >>> Now that I have made everything clear as mud, I also want to point >>> out that we should consider modifying our example. I believe this, >>> due simply to the fact that we are starting with a web application >>> within our web services model, and I think implementers will >>> conclude that our standard is only for web applications in web >>> services. I feel more confident in saying this because I have worked >>> with software engineers and software engineering executives, and a >>> more literal-minded group would be hard to find. So, while I am not >>> going to do more than leave the issue open, I do think we better >>> consider starting out with a portlet that just delivers lists of >>> stocks and prices, and maybe some analyst reports, and add the >>> application as the more advanced element later, since it requires >>> actual remote manipulation of the data, if the app doesn't live on >>> the client machines. (Which I would assume from the current >>> scenario, that it does not.) >>> >>> Ciao, >>> Rex >>> >>> At 12:41 PM -0400 10/24/03, Rich Thompson wrote: >>> While I appreciate the concern that the primer not reuse such an old >>> example that people wonder what is new or useful here, it is also >>> true that a familiar base example keeps understanding the example >>> from distracting from what it is supposedly demonstrating. Perhaps a >>> useful middle ground is to use a familiar (even overly familiar) >>> example for walking through the interfaces and then double back for >>> a "More advance, real world example" in a section of its own. This >>> provides a simple learning ground and then walks the reader through >>> the process of designing a realistically complicated scenario. >>> Comments? >>> >>> Rich >>> *Rex Brooks <rexb@starbourne.com>* >>> 10/24/2003 12:44 PM >>> To >>> wsrp-primer@lists.oasis-open.org, wsrp-markup@lists.oasis-open.org >>> cc >>> Subject >>> [wsrp-primer] A consistent use-case for primer >>> >>> >>> >>> >>> >>> Hi Folks, >>> >>> After our primer telecon yesterday, when I hastily withdrew my >>> objection to using a stock brokerage/portfolio management example, >>> for various reasons I don't want to belabor in this message, I got to >>> thinking about it some more and decided I really don't think that it >>> is a wise choice after all. I have been banging my head against all >>> the older scenarios, and was reminded in most of them that there were >>> problems here and there with most, as well as with my own current >>> project, building a presentation of WSRP and CAP (Common Alerting >>> Protocol) from the Emergency Management TC. >>> >>> One of my concerns was that it would be best to have one example >>> follow through the tutorials we put into the primer, and that this >>> example be used for demonstrating the publish, find, bind features in >>> the Registration Interface, including use of both UDDI and ebXML, in >>> keeping with our plan to offer real-world examples and code (examples >>> and code that would work, that is, not the actual Portals or Services >>> of the example). It must also fit with demonstrations of the Service >>> Description Interface, Markup Interface and Portlet Management >>> Interface, all of which it does. >>> >>> So, while going over the primer thus far and continuing to review the >>> use cases and scenarios we have dealt with in the process of building >>> WSRP 1.0 and moving now into issues for 1.1, I decided that I could >>> offer an alternate example by adapting the "Search UDDI..." scenario >>> Andre worked up and which is available in the documents section of >>> the WSRP webpages. >>> >>> However, before I actually include it in my edit of the current >>> primer and start asking for assistance in building some sample >>> portlets and a Portal Design exercise as part of the Markup >>> Interface, I wanted to share this idea with you all and ask if it >>> makes sense to you and if you would be willing to use it. >>> >>> From our Scenario Form Description: HealthCareConcerns, LLC., is a >>> Portal development company, specializing in horizontal Medical >>> Information Resources (Medical Informatics), including a searchable >>> database of insurance providers, and vertical Medical Supplies >>> procurement/supply chains. It develops a broad range of >>> Healthcare-related Portals using domain-specific expertise from >>> related but unintegrated market segments in order to focus on making >>> itself a Healthcare Information Clearinghouse. >>> >>> HealthCareConcerns, ILLC., designs and builds specialized sites for >>> knowledge-base delivery clients, by integrating domain-specific >>> Portlets into Portals that service a range of information providers >>> from specific niche markets to more broadly based Heathcare market >>> segments, such a Home-based Elderly Care Products and Services or >>> Maternity-related products and services. It will use UDDI and ebXML >>> to publish its services and the services of its Portal clients and to >>> search for new Portlets, based on an initial profile it makes of its >>> specific clients. >>> >>> While some of the expertise and knowledge I have developed over the >>> last several years can be used in this, it is not strictly an >>> instance of the proof of concept Public Healthcare Preparedness >>> Portal I am working on. However, I won't say that it doesn't also fit >>> into other plans I have, but I always try to make any project serve >>> more than a single purpose if I can. >>> >>> This scenario also provides what I think is a decent range of the >>> kinds of Portals and Portlets WSRP can facilitate without either >>> relying on either well-used examples such as stock market/brokerages >>> or purchasing orders. Of all the previous scenarios the one I would >>> have chosen as both simple and fairly straight forward was the memory >>> configurator, but it has a couple of problems, mostly due to its >>> being a customization example, which is a bit ahead of 1.0. >>> >>> Please let me know what you think. I am asking the markup committee, >>> too since this is probably where I will be asking for samples >>> regardless of which scenario we choose to follow through with for the >>> Tutorials. >>> >>> Ciao, >>> Rex >>> -- >>> Rex Brooks >>> GeoAddress: 1361-A Addison, Berkeley, CA, 94702 USA, Earth >>> W3Address: http://www.starbourne.com >>> Email: rexb@starbourne.com >>> Tel: 510-849-2309 >>> Fax: By Request >>> >>> >>> -- >>> >>> Rex Brooks >>> GeoAddress: 1361-A Addison, Berkeley, CA, 94702 USA, Earth >>> W3Address: http://www.starbourne.com >>> Email: rexb@starbourne.com >>> Tel: 510-849-2309 >>> Fax: By Request >> > >
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]